The Prediction of Optimal Torrefaction Condition Palm Kernel Shell
based on Elemental Composition
Karelius
1
, Made Dirgantara
1
, Nyahu Rumbang
1
, Komang Gde Suastika
1
, I Dewa Gede Putra
Prabawa
2
and Lusi Ernawati
3
1
University of Palangka Raya
2
Industry Research and Standardization Center, Ministry of Industry, 02 Panglima Batur Barat street, Banjarbaru,
Indonesia
3
Institut Teknologi Kalimantan
Keywords: Elemental Composition, Energy Yield, High Heating Value, Palm Kernel Shell, Torrefaction
Abstract: Utilization of biomass as an energy source is very important, especially by-products from forestry, plantations,
and livestock. Biomass as fuel still has many disadvantages, both in terms of water content, calories, and
weight. To overcome this, a thermochemical process is needed, namely Torrefaction. Torefaction is a heating
process between 200-325°C with minimal oxygen conditions. This process is being widely used to improve
the properties of biomass as fuel and reduce the weaknesses of biomass, such as low heating and energy
density, high inertia, low combustion efficiency, and high milling energy. In this study of palm kernel shell
torrefaction was carried out from 0.5 cm sieve with three variations in temperature, and residence time follows
250 0C - 300 0C and 20-40 minutes then analyzed ultimate. The high heating value is predicted using ultime
data calculated by the equation of Chang Dong and Azevedo, Fried, and Özyuğuran then used to determine
energy yield. The optimum process for torrefaction of palm kernel shell according to calorific value and
energy yield is 275 0C with residence time 20 minutes.
1 INTRODUCTION
Utilization of biomass as an energy source is very
important, especially by-products from forestry,
plantations and livestock (Mafu et al. 2016;
Mamvura, Pahla, and Muzenda 2018; Mohd Faizal et
al. 2018). Based on the Intergovernmental Panel on
Climate Change (IPCC) declaration on the Climate
Change Synthesis Report indicated 25% of
greenhouse gases came from the electricity and heat
industry, where 40% came from the coal industry.
Under these conditions, potential raw materials that
are environmentally friendly and sustainable and can
be integrated with steam power plants are biomass.
Compared to other environmentally friendly energy
such as solar and wind cells, biomass has advantages
instability and can be controlled according to needs.
Biomass as fuel still has many disadvantages, both
in terms of water content, calories, and weight. To
overcome this, a thermochemical process is needed,
namely Torrefaction (Z.c, H.j, and D.m.j 2017; Barta-
Rajnai et al. 2017; Pahla et al. 2017). Torrefaction is
a heating process between 200-325°C with minimal
oxygen conditions. This process is being widely used
to improve the properties of biomass as fuel and
reduce the weaknesses of biomass, such as low
heating and energy density, high inertia, low
combustion efficiency, and high milling energy
(Thaim and Rasid 2016; Bach, Skreiberg, and Lee
2017).
The heating value of biomass is an important
parameter for the planning and the control of power
plants using this type of fuel (Friedl et al. 2005).
Optimizing torrefaction parameters leads to an
improvement in both the quantity and quality of the
torrefied biomass. With that in mind, it means that
maximizing biomass higher heating value whilst
minimizing biomass weight loss from its raw state
results in optimized torrefaction parameters
(Mamvura, Pahla, and Muzenda 2018). Thus, the
focus was on finding the optimized torrefaction
parameters, which has an energy yield of more than
100%.
In this study of palm kernel shell torrefaction was
carried out from 0.5 cm sieve with three variations in
46
Karelius, ., Dirgantara, M., Rumbang, N., Suastika, K., Prabawa, I. and Ernawati, L.
The Prediction of Optimal Torrefaction Condition Palm Kernel Shell based on Elemental Composition.
DOI: 10.5220/0009405000460051
In Proceedings of the 1st International Conference on Industrial Technology (ICONIT 2019), pages 46-51
ISBN: 978-989-758-434-3
Copyright
c
 2020 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

temperature, and residence time follows 250
0
C - 300
0
C and 20-40 minutes then analyzed ultimate. The
high heating value (HHV) is predicted using ultimate
data calculated by the equation of Chang Dong and
Azevedo, Fried, and Özyuğuran then used to
determine energy yield.
2 METHODS
2.1 Preparation and Torrefaction
Palm kernel shells dried in the sun for 3 days to
reduce the moisture content. The feedstock size is
reduced to a size of ± 0.5 cm. Torrefaction process is
then performed. The torrefaction process is shown in
Figure 1, with an inert condition (Thaim and Rasid
2016), the torrefaction temperature varied from
250°C, 275°C, 300°C (Pahla et al. 2017), with
residence time varied 20, 30 dan 40 minutes
(Mamvura, Pahla, and Muzenda 2018; Pahla et al.
2017; Pahla, Ntuli, and Muzenda 2018).
Figure 1: Schematic of Terrefaction Proces (Mamvura,
Pahla, and Muzenda 2018)
2.2 Ultimate Analysis
The ultimate analysis is done using the TruSpec CHN
tool from Leco Inc. based on ASTM D 5373-16
standards (to determine the composition of carbon,
hydrogen, and nitrogen). The proximate and ultimate
analysis is carried out on samples before treatment
and after the torrefaction on the process. (Mohd
Faizal et al. 2018; Susanty, Helwani, and Zulfansyah
’ 2016; Nyoman Sukarta and Sri Ayuni 2016)
2.3 Prediction HHV and Energy Yield
2.3.1 Prediction of HHV
The high heating value is predicted using ultime data
calculated using following the equation of Chang
Dong and Azevedo, Fried, and Özyuğuran:
(Chang Dong and Azevedo 2005)
𝐻𝐻𝑉𝑀𝐽/𝐾𝑔  1.3675  0.3137𝐶  0.7009𝐻
 0.0318𝑂
(Friedl et al. 2005)
𝐻𝑉

𝑀𝐽/𝐾𝑔

 0.00355𝐶

 0.232𝐻  2.23𝐻 
0.0512𝐶𝐻  0,131𝑁 20.6
(Özyuğuran, Yaman, and Küçükbayrak 2018)
𝐻𝐻𝑉

𝑀𝐽/𝐾𝑔

 4.6246  0.2732𝑁  0.4120𝐶
 0.5992𝐻  0.01841𝑂
Where: C = percentage of carbon; H = percentage
of hydrogen; N= percentage of nitrogen; O =
percentage of Oxygen
2.3.1 Prediction of Energy Yield
Two parameters are used to determine the success of
the torrefaction process are mass yield and energy
yield. The focus in the torrefaction process is to get a
high HHV value and low mass loss due to this process
(Lau et al. 2018; Mohd Faizal et al. 2018; Rodrigues,
Loureiro, and Nunes 2018).
1) Mass yield computed as
𝑀𝑎𝑠𝑠 𝑦𝑖𝑒𝑙𝑑 𝑀𝑌 
𝑀

𝑀

 100%
Where M
r
is mass of raw biomass and M
t
is mass
of torrefied biomass.
2) Energy yield was computed as
𝐸𝑛𝑒𝑟𝑔𝑦 𝑦𝑖𝑒𝑙𝑑  𝑀𝑌 
𝐶𝑉

𝐶𝑉

Where CV
t
and CV
P
is the calorific value of the
biomass after and before torrefaction.
3 RESULTS AND DISCUSSION
3.1 Ultimate Analysis and Prediction
HHV Value
Palm Kernel shell was ultimately analyzed before and
after torrefaction to obtain the elemental composition
and predict the HHV value. An increase in carbon and
hydrogen, while the decrease in nitrogen and oxygen
levels of palm shell were obtained after torrefaction,
as shown in Table 1.
It was known that hydrogen (H
2
) and oxygen (O
2
)
are the main components in biomass that most lost
during the torrefaction process. Oxygen (O
2
) is the
main component in the formation of the volatile
The Prediction of Optimal Torrefaction Condition Palm Kernel Shell based on Elemental Composition
47

compound, which can produce an abundant phenol at
a temperature range of 200-300
o
C. Whereas
hydrogen (H
2
) is the main component of liquid smoke
(condensate) constituents as a result of the
torrefaction process. It has been reported that the ratio
of carbon (C) to hydrogen (H
2
) will increase and
volatile levels will decrease with reducing hydrogen
and oxygen levels which is indicated that the high
quality of fuels has low O/C and H/C (Chen et al.,
2017).
The ratio of O/C has a large influence on the
heating value of materials. Pahla et al. reported that a
decrease in the O/C ratio would result in an increase
in heating value. When the ratio of O/C was so huge,
the chemical energy was also relatively high, thereby
reduce in gasification efficiency due to excessive
oxidation. Moreover, in the combustion process,
carbonization causes increasing carbon and hydrogen
content and decreasing oxygen content. Therefore,
torrefaction plays an important role in obtaining the
low heating value of palm shell.
As shown in Table 1, the highest carbon value was
ultimately obtained at a temperature of 275
o
C, and the
oxygen was significantly decreased causes the O/C
ratio is lower than other temperatures. The prediction
of HHV values based on three different equations also
shows that the HHV value at 275
o
C is greater than the
others. The holding time of 30 minutes has the lowest
H/C and O/C ratio and the highest HHV value. This
result indicates that the temperature of 275
o
C was
used to get the optimum time of the torrefaction
process.
There was a difference in the results of the final
analysis between the effect of temperature and
residence time. Whereat temperature of 275
o
C and
residence time 40 minutes the ratio of H / C and O /
C is higher compared to the conditions during time
variation. This is due to different water levels at the
end of the torrefaction process, which affects the
calculation of O.
Table 1: Elemental composition and prediction of HHV value
Condition of
Torrefaction
N (%) C (%) H (%) O (%) O/C H/C
HHV (MJ/Kg)
Chang-
Dong and
Azevedo
Fried Özyuğuran
Control 0,42 36,69 3,74 47,4 1,29 0,10 14,27 15,61 13,72
Temperature
effect (40
minutes resident
time)
250 °C 1,05 44,93 5,54 44,05 0,98 0,12 18,01 17,87 18,31
275 °C 0,98 52,38 5,58 37,79 0,72 0,11 20,18 20,84 21,26
300 °C 1,10 52,01 5,32 38,91 0,75 0,10 19,91 20,58 21,01
Resident time
effect
(temperature
275
o
C)
20 minute 1,29 63,43 4,99 26,49 0,42 0,08 22,87 25,42 25,34
30 minute 1,32 65,47 4,87 24,83 0,38 0,07 23,37 26,26 26,08
40 minute 1,20 61,61 4,62 29,30 0,48 0,07 22,13 24,21 24,39
ICONIT 2019 - International Conference on Industrial Technology
48

3.2 Prediction of Energy Yield
To determine the optimal conditions of the
torrefaction process not only based on calorific values
but also must consider the mass lost or mass yield of
the process (Mamvura, Pahla, and Muzenda 2018; da
Silva et al. 2018). Energy yield is a parameter that is
influenced by the value of HHV and also mass yield.
If the value of energy yield is above 1 or 100%,
the torrefaction process has been able to increase the
energy that can be used in biomass. Based on the
HHV values, the energy yield results are shown in
Table 2, Table 3, and Table 4.
Table 2: Energy yield prediction base on Chang-Dong and Azevedo equation
Condition of
Torrefaction
Mo (g) Mt (g)
Cvi
(MJ/Kg)
CVt
(MJ/Kg)
Weigh
Loss
Mass
Yield
Energy
Yield
Temperature
effect (40 minutes
resident time
)
250 °C 500 371 14,27 18,01 25,80 74,20 0,94
275 °C 500 313 14,27 20,18 37,40 62,60 0,89
300 °C 500 236 14,27 19,91 52,80 47,20 0,66
Resident time
effect
(temperature
275
o
C
)
20 minute 500 327 14,27 22,87 34,60 65,40 1,05
30 minute 500 295 14,27 23,37 41,00 59,00 0,97
40 minute 500 271 14,27 22,13 45,80 54,20 0,84
Table 3: Energy yield prediction base on Fried equation
Condition of
Torrefaction
Mo (g) Mt (g)
Cvi
(MJ/Kg)
CVt
(MJ/Kg)
Weigh
Lost
Mass
Yield
Energy
Yield
Temperature
effect (40 minutes
resident time
)
250 °C 500 371 15,61 17,87 25,80 74,20 0,85
275 °C 500 313 15,61 20,84 37,40 62,60 0,84
300 °C 500 236 15,61 20,58 52,80 47,20 0,62
Resident time
effect
(temperature 275
o
C
)
20 minute 500 327 15,61 25,42 34,60 65,40 1,06
30 minute 500 295 15,61 26,26 41,00 59,00 0,99
40 minute 500 271 15,61 24,21 45,80 54,20 0,84
The Prediction of Optimal Torrefaction Condition Palm Kernel Shell based on Elemental Composition
49

Generally, based on the HHV values of the three
equations used to show similar results. Based on the
results of temperature variations, the highest energy
yield value is at 250
o
C and continues to decline with
increasing temperature. This is caused by the length
of residence time, which causes a large mass loss. In
these conditions, there is no energy yield value that
exceeds 1, meaning that the process of torrefaction
under these conditions cannot increase the amount of
energy that can be used. A different result is shown
by energy yield with the variation of time at a
temperature of 275
o
C with a residence time of 20
minutes at the torrefaction process can produce
energy yields bigger than 1, which means that under
these conditions the torrefaction process can increase
the value of calories and energy that can be used.
4 CONCLUSIONS
The torrefaction process is not only concerned with
the value of HHV obtained but also causes the mass
lost. Based on the results of HHV value and energy
yield, the optimum conditions were achieved using
the Batch method with a temperature of 275
o
C and
the residence time of 20 minutes. In this condition,
the torrefaction process succeeded in reducing the
O/C and H/C ratios and increasing the fuel energy of
the palm kernel shell.
ACKNOWLEDGMENTS
This work was supported by Badan Pengelola Dana
Perkebunan Kelapa Sawit (BPDPKS) who have
funded through the Grant Palm Research Scheme.
REFERENCES
Bach, Quang-Vu, Øyvind Skreiberg, and Chul-Jin Lee.
2017. “Process Modeling and Optimization for
Torrefaction of Forest Residues.” Energy (November)
https://doi.org/10.1016/j.energy.2017.07.040.
Barta-Rajnai, E., L. Wang, Z. Sebestyén, Zs. Barta, R.
Khalil, Ø. Skreiberg, M. Grønli, E. Jakab, and Z.
Czégény. 2017. “Effect of Temperature and Duration of
Torrefaction on the Thermal Behavior of Stem Wood,
Bark, and Stump of Spruce.” Energy Procedia, 8th
International Conference on Applied Energy,
ICAE2016, 8-11 October 2016, Beijing, China, 105
(May): 551–56.
https://doi.org/10.1016/j.egypro.2017.03.355.
ChangDong, Sheng, and J. L. T. Azevedo. 2005.
“Estimating the Higher Heating Value of Biomass
Fuels from Basic Analysis Data.” Biomass and
Bioenergy 28 (5): 499–507.
Chen, Yun-Chun, Wei-Hsin Chen, Bo-Jhih Lin, Jo-Shu
Chang, and Hwai Chyuan Ong. 2017. “Fuel Property
Variation of Biomass Undergoing Torrefaction.”
Energy Procedia, 8th International Conference on
Applied Energy, ICAE2016, 8-11 October 2016,
Beijing, China, 105 (May): 108–12.
https://doi.org/10.1016/j.egypro.2017.03.287.
Table 4: Energy yield prediction base on Özyuğuran equation
Condition of
Torrefaction
Mo (g) Mt (g)
Cvi
(MJ/Kg)
CVt
(MJ/Kg)
Weigh
Loss
Mass
Yield
Energy
Yield
Temperature
effect (40 minutes
resident time)
250 °C 500 371 13,72 18,31 25,80 74,20 0,99
275 °C 500 313 13,72 21,26 37,40 62,60 0,97
300 °C 500 236 13,72 21,01 52,80 47,20 0,72
Resident time
effect
(temperature
275
o
C)
20 minute 500 327 13,72 25,34 34,60 65,40 1,21
30 minute 500 295 13,72 26,08 41,00 59,00 1,12
40 minute 500 271 13,72 24,39 45,80 54,20 0,96
ICONIT 2019 - International Conference on Industrial Technology
50

Friedl, A., E. Padouvas, H. Rotter, and K. Varmuza. 2005.
“Prediction of Heating Values of Biomass Fuel from
Elemental Composition.” Analytica Chimica Acta,
Papers Presented at the 9th International Conference on
Chemometrics in Analytical Chemistry, 544 (1): 191–
98. https://doi.org/10.1016/j.aca.2005.01.041.
Lau, Hun Shen, Hoon Kiat Ng, Suyin Gan, and Seyed
Amirmostafa Jourabchi. 2018. “Torrefaction of Oil
Palm Fronds for Co-Firing in Coal Power Plants.”
Energy Procedia, Special Issue of the Fourth
International Symposium on Hydrogen Energy,
Renewable Energy and Materials, 2018 (HEREM
2018), 144 (July): 75–81.
https://doi.org/10.1016/j.egypro.2018.06.010.
Mafu, Lihle D., Hein W. J. P. Neomagus, Raymond C.
Everson, Marion Carrier, Christien A. Strydom, and
John R. Bunt. 2016. “Structural and Chemical
Modifications of Typical South African Biomasses
during Torrefaction.” Bioresource Technology 202
(February): 192–97.
https://doi.org/10.1016/j.biortech.2015.12.007.
Mamvura, T. A., G. Pahla, and E. Muzenda. 2018.
“Torrefaction of Waste Biomass for Application in
Energy Production in South Africa.” South African
Journal of Chemical Engineering 25 (June): 1–12.
https://doi.org/10.1016/j.sajce.2017.11.003.
Mohd Faizal, Hasan, Hielfarith Suffri Shamsuddin, M.
Harif M. Heiree, Mohd Fuad Muhammad Ariff Hanaffi,
Mohd Rosdzimin Abdul Rahman, Md. Mizanur
Rahman, and Z. A. Latiff. 2018. “Torrefaction of
Densified Mesocarp Fibre and Palm Kernel Shell.”
Renewable Energy 122 (July): 419–28.
https://doi.org/10.1016/j.renene.2018.01.118.
Nyoman Sukarta, I, and Putu Sri Ayuni. 2016. “analisis
proksimat dan nilai kalor pada pellet biosolid yang
dikombinasikan dengan biomassa limbah bambu.” JST
(Jurnal Sains Dan Teknologi) 5 (August).
https://doi.org/10.23887/jst-undiksha.v5i1.8278.
Özyuğuran, Ayşe, Serdar Yaman, and Sadriye
Küçükbayrak. 2018. “Prediction of Calorific Value of
Biomass Based on Elemental Analysis,” 7.
Pahla, G., T. A. Mamvura, F. Ntuli, and E. Muzenda. 2017.
“Energy Densification of Animal Waste Lignocellulose
Biomass and Raw Biomass.” South African Journal of
Chemical Engineering 24 (December): 168–75.
https://doi.org/10.1016/j.sajce.2017.10.004.
Pahla, G., F. Ntuli, and E. Muzenda. 2018. “Torrefaction of
Landfill Food Waste for Possible Application in
Biomass Co-Firing.” Waste Management 71 (January):
512–20.
https://doi.org/10.1016/j.wasman.2017.10.035.
Rodrigues, A., L. Loureiro, and L. J. R. Nunes. 2018.
“Torrefaction of Woody Biomasses from Poplar SRC
and Portuguese Roundwood: Properties of Torrefied
Products.” Biomass and Bioenergy 108 (January): 55–
65. https://doi.org/10.1016/j.biombioe.2017.11.005.
Silva, Carlos Miguel Simões da, Angélica de Cássia
Oliveira Carneiro, Benedito Rocha Vital, Clarissa
Gusmão Figueiró, Lucas de Freitas Fialho, Mateus
Alves de Magalhães, Amélia Guimarães Carvalho, and
Welliton Lelis Cândido. 2018. “Biomass Torrefaction
for Energy Purposes – Definitions and an Overview of
Challenges and Opportunities in Brazil.” Renewable
and Sustainable Energy Reviews 82 (February): 2426–
32. https://doi.org/10.1016/j.rser.2017.08.095.
Susanty, Wenny ’, Zuchra ’ Helwani, and Zulfansyah ’.
2016. “Torefaksi Pelepah Sawit : Pengaruh Kondisi
Proses terhadap Nilai Kalor Produk Torefaksi.” Jurnal
Online Mahasiswa Fakultas Teknik Universitas Riau 3
(1): 1–6.
Thaim, Thuraiya, and Ruwaida Abdul Rasid. 2016.
“Improvement Empty Fruit Bunch Properties through
Torrefaction.” Australian Journal of Basic and Applied
Sciences, 8.
Z.c, Bourgonje, Veringa H.j, and Smeulders D.m.j. 2017.
“The New Method to Characterize the Gas Emissions
during Torrefaction Real-Time.” Fuel Processing
Technology 164: 24–32.
The Prediction of Optimal Torrefaction Condition Palm Kernel Shell based on Elemental Composition
51